This text has been translated from Norwegian with the assistance of GPT UiO.
How is your research put to use?
– The projects of mine that have gained the most attention and have been used outside the research sector concern the prediction of election outcomes. I have two projects in this area: one is a daily updated website that predicts results in parliamentary elections based on opinion polls. I launched this before the 2021 parliamentary election, and it has been cited by NRK Dagsrevyen, Politisk kvarter and TV 2 Nyhetskanalen, among others. Traffic to the site was high both before the 2021 election and this year’s election.
– The other project is a collaboration with NRK, where I developed a forecasting model for use on television on election night. The need for such a model arises because the running shares of counted votes can differ quite substantially from the final results, giving a misleading picture early in the evening. In addition, NRK and other media often wish to determine various results before 100% of the votes have been counted. The model I created for NRK enabled them to draw a wide range of conclusions throughout the evening — including whether the parties on the left or right had secured a majority. The model also proved to be more accurate than all previous forecasts ever shown on Norwegian TV.
Who used the results - for instance in policy, in practice, in private sector, public administration or civil society?
– The results from these projects have been cited in most of the major media outlets in Norway. I have received feedback that the forecasting model I built for NRK is extremely valuable for their election coverage. I also know that other media outlets are interested in the model.
– In addition, my website has seen heavy traffic from ordinary citizens and politicians. Before the 2021 election, I was contacted by sitting members of parliament and aspiring candidates who were eager to know how the election might turn out, and who used the website to put isolated opinion polls into perspective.
How have you worked to ensure that your research contributes to a positive social development?
– In the past, I have researched many different topics that I believe are relevant for decision-makers who want positive social development. For example, I have sought to shed light on the extent to which the EU’s decision-making processes are transparent and efficient, which can reveal potential for improvement. As for election prediction, I would not claim that this is crucial for societal development, although good predictions may help maintain public trust in how we run elections and count votes.
– I see my most important role as communicating knowledge about research methods and statistics, through teaching, answering questions from the media, and developing models in collaboration with others — such as NRK. I believe there is considerable scope to raise the public’s understanding of research and statistical analysis, which I think can improve the quality of public debate and lead to better decisions across many areas of society. I have tried to contribute to this both through the courses I teach and by challenging questionable claims in the media — whether they concern causal inferences or interpretations of opinion polls.
Do you have any tips or experiences you want to share with colleagues, so that their research is put to use?
– My experiences may not be transferable to other researchers’ situations, but clearly the chances that what you create will be used outside the research sector increases if you aim to do something relevant. There can be a certain tension between focusing on relevance and conducting basic research, so not everyone should dwell on this too much — but I think it is positive if some of us do, at least part of the time.
– The projects of mine that have been most widely used outside the research sector are unorthodox in the sense that they are not typical research projects and will not necessarily lead to published research articles. These projects have had different objectives — the website I created began as a sort of hobby project aimed at presenting results of public interest, while the project with NRK’s forecasting model began at NRK’s initiative with the goal of using the results on TV.
– At the same time, these projects have been made possible by more traditional research projects. The articles I am working on today are often purely methodological, in which I develop statistical models and associated software. The hope is, of course, that many researchers will find this useful, even if the eventual reach may be narrow. An important side-effect of working on such narrowly focused projects is that I acquire skills that allow me to carry out more applied projects, and I try to find a balance between typical academic projects and more outward-facing activities. I believe there will be considerable variation in what the right balance is for different researchers in different fields, so I think everyone needs to find the balance that works for them.